Nvm-based method for performance acceleration of containers

Abstract

The present disclosure discloses a NVM-based method for performance acceleration of containers. The method comprises classifying each image layer of mirror images as either an LAL (Layer above LDL) or an LBL (Layer below LDL) during deployment of containers; storing the LALs into a non-volatile memory and selectively storing each said LBL into one of the non-volatile memory and a hard drive; acquiring hot image files required by the containers during startup and/or operation of the containers and storing the hot image files required by the containers into the non-volatile memory; and sorting the mirror images in terms of access frequency according to at least numbers of times of access to the hot image files so as to release the non-volatile memory currently occupied by the mirror image having the lowest access frequency when the non-volatile memory is short of storage space.

Description

FIELD[0001]The present invention relates to the field of virtualization, and more particularly to optimization of Docker in an NVM-based virtualization environment so as to accelerate deployment and startup of containers with reduced writing wear and usage to NVMs (Non-Volatile Memories), and even more particularly to an NVM-based method for performance acceleration of containers.DESCRIPTION OF RELATED ART[0002]After the advent of virtualization, container-related technologies have gradually become the most significant, revolutionary technologies in the field of cloud computation. As a trend, more and more effort, research, development and promotion have been made for container-related technologies and their applications in recent years, making these technologies grow rapidly and become a major part of cloud computation. Container-based virtualization is a lightweight method that facilitates resource isolation and management in Linux environments. The use of containers is similar to...

AI Technical Summary

Benefits of technology

[0013]To address the shortcomings of the prior art, the present invention provides an NVM-based performance acceleration method for containers. The present invention, according to the traits of container deployment and startup and the good properties of NVMs, accelerates container deployment and startup by storing a part of mirror images to an NVM, with reduced writing wear and usage to the NVM, so as to address the problem about slow container deployment and startup seen in the traditional architecture due to restrictions caused by disk I / O. What is crucial is that the present invention enhances container performance by improving the storage architecture, so as to significantly accelerate deployment and startup of containers built based on mirror images of any version.

Problems solved by technology

However, slowness of disk I / O brings limitation to container deployment and startup.

Due to the three features, in the process of pulling mirror images, when all the mirror images have been fully downloaded, there will be a large part of mirror images waiting to be stored in the hard drive, resulting in very high latency and I / O burst.

Containerization is in nature a process of enjoying virtualization of all resources, but in practice the slowness of disk I / O delays startup of containers.

NVMs have limited write endurance and much smaller capacity as compared to disks, making it necessary to combine NVM and Docker reasonably and effectively.

However, the method in fact requires the costs of reduced compatibility and limited yield.

In view of this, the known method is not as effective as asserted and has numerous restrictions.

Images